NASA Goddard Space Flight Center
Experiment 5

Home Up Basic Info/FAQ Other Metal Whiskers Literature References Whisker Failures Whisker Anecdotes Photo Gallery Video Gallery Related Links What's New Contact Us


Experiment #5: Last Updated October 2002 -  Continued Monitoring Planned 
Tin Whiskers Growing from the Terminations of Pure Tin Plated Ceramic Chip Capacitors After Temp Cycle Followed By Ambient Storage

hybrid-chip-caps.jpg (83139 bytes)
 Inside of a Hybrid Microcircuit Package with Pure Tin-Plated Ceramic Chip Capacitors
Mounted by Conductive (Silver) Epoxy

2002oct-chip-cap03-01.jpg (578847 bytes)
Tin Whisker Growth on Capacitor Terminations after 200+ Temp Cycles 
From -40C to +90C Followed by Months of Ambient Storage (2002-Oct)

Interim Observations
Prior Research

This experiment documents and monitors tin whisker growth on one lot of pure tin-plated multilayer ceramic capacitors (MLCCs) mounted inside a hybrid using conductive (silver) epoxy and then subjected to extensive thermal cycling followed by long term ambient storage.  The experiment was initiated in 2001 by a non-NASA entity who then donated one of their test samples to NASA Goddard tin whisker researchers for continued monitoring.  Details of this study and interim observations were formally published in June 2002 at the American Electroplaters and Surface Finishers (AESF) SUR/FIN conference in Chicago, IL.  The paper also discusses similar but unrelated experiments by others attempting to induce whisker growth on tin-plated MLCCs.

  1. J. Brusse, "Tin Whisker Observations on Pure Tin Plated Ceramic Chip Capacitors", AESF SUR/FIN Conference, June 2002, pp 45-61
  2. Slide presentation is also available.

The NASA Goddard Whisker WWW Site will be used to provide updates as monitoring of these specimens continues.


In 2001 a hybrid microcircuit manufacturer informed NASA Goddard tin whisker researchers of their surprising observation of profuse whisker formation on pure tin-plated ceramic chip capacitors (MLCCs).  The hybrid manufacturer made this discovery during an evaluation intended to assess the electrical and mechanical integrity of joints made between pure tin-plated MLCC terminations and gold-plated substrate pads when using conductive (silver) epoxy instead of solder reflow for mounting.  Their experiments included the following basic approach:

Sample Preparation:  Conductive epoxy mount pure tin-plated MLCCs to gold plated termination pads inside of hermetically sealed hybrids
Subject sealed hybrids to:
Condition 1: Temperature Cycling  -40C to +90C for up to 500 cycles
Condition 2: High Temperature Storage (+90C) for 400 hours
Initial Results:
Condition 1:  Whiskers up to 100 um long observed after as little as 200 cycles
Condition 2:  No whiskers found

In May 2001 the hybrid manufacturer donated one thermally cycled evaluation hybrid package (200+ cycles per Condition 1 above) to NASA Goddard for continued observation.  The package contained 6 pure tin-plated ceramic chip capacitors (size 0805) all reported to be from one manufacturing lot.  As received by NASA Goddard, the hybrid package had already been delidded.  The manufacturer supplied photos of the capacitor terminations to illustrate the extent of whisker growth they observed after temp cycle.  Upon receipt at NASA Goddard the sample was reinspected and the whisker formation was confirmed to be consistent with the extent of growth reported by the hybrid manufacturer with all 6 capacitors showing clear evidence of tin whisker formation.   NASA Goddard has since retained this sample in office ambient storage for ongoing observation and analysis.

NOTE:  The hybrid manufacturer reports they originally intended to use palladium/silver terminated capacitors in an application where the capacitors were to be mounted to a substrate using conductive epoxy.  Erroneously, the supplier of the capacitors provided pure tin plated capacitors (standard commercial offering) to fill the user's order.  CAVEAT EMPTOR.  The hybrid manufacturer noticed a visible difference in appearance of the termination finish at incoming inspection.  The subsequent evaluations were initiated to determine if the mistaken shipment of capacitors could be used for production anyway.

Interim (Summary) Findings (reverse chronological order):

October 2002:  Additional SEM Photos indicate the longest whiskers observed in Dec. 2001 continue to show no further growth.  No attempts were made to characterize whether any new whiskers had formed nor whether "shorter" whiskers showed any continued growth since Dec. 2001
May 2002:  High Power SEM Photos were taken document the length of the longest whisker and to show that many of the whiskers possess regularly-spaced circumferential "rings" in addition to the commonly reported longitudinal striations.
December 2001: Some of the tin whiskers on the MLCCs have grown to greater than 200 microns long after 6 - 8 months of ambient storage that followed the initial thermal cycle experiments done by the hybrid manufacturer.  Note: Maximum length reported immediately after Thermal Cycle was ~ 100 microns.  This observation shows that the whiskers have continued to grow despite the discontinuation of the thermal cycle environment which "initiated" the growth process. 
October 2001:  GSFC removes one capacitor from the hybrid for cross section/SEM/EDS analysis (see GSFC Materials Branch Analysis Report).  Analysis reveals capacitor termination structure is "typical" for this product type and contains very similar attributes previously reported to be beneficial for reducing whisker propensity:.
Tin layer ~6.5 microns thick
Nickel barrier layer ~6.5 microns thick
Silver frit ~17 microns thick
Barium titanate ceramic body.
August 2001:  GSFC performs receiving inspection (SEM/optical) of a hybrid used in the manufacturer's thermal cycle evaluation (> 200 cycles between -40C and +90C).  As received, the hybrid had been delidded with the capacitors still mounted on the substrate (conductive epoxy).  GSFC inspection confirms information reported by the hybrid manufacturer immediately following thermal cycle testing (max. whisker length is ~ 100 microns)
~February 2001:  Original SEM photos taken by hybrid manufacturer immediately after >200 thermal cycles between -40C to +90C of hybrid package containing 6 pure tin-plated ceramic chip capacitors conductive epoxy mounted to gold-plated termination pads .  Hybrid manufacturer reports very dense tin whisker formation with max. whisker lengths on the order of 100 microns.

October 2002 - Photos Below are from Same Capacitors T-cycled in early 2001 then stored in Ambient Office Conditions
Courtesy NASA Goddard Space Flight Center Parts Analysis Lab

2002oct-chip-cap03-01.jpg (578847 bytes) 2002oct-chip-cap05-06.jpg (775524 bytes) 2002oct-chip-cap05-04.jpg (411855 bytes)
2002oct-chip-cap03-02.jpg (378545 bytes) 2002oct-chip-cap05-02.jpg (378786 bytes) 2002oct-chip-cap01-02.jpg (453013 bytes)

May 2002- High Power SEM Images to Show How SOME of the Whiskers have Circumferential Rings
Courtesy of NASA Goddard Space Flight Center

cap5-whisker1-238micron-overall.jpg (650687 bytes)
Longest Whisker Found ~ 240 microns

cap1-whisker1-overall.jpg (478361 bytes)
Whisker "#1" - Overall View
cap1-whisker1-base02.jpg (471248 bytes)
Whisker "#1" - Base of Whisker
cap1-whisker1-mid02.jpg (426862 bytes)
Whisker "#1" - Middle of Whisker (Notice "Rings")
cap1-whisker1-tip.jpg (433556 bytes)
Whisker "#1" - Tip of Whisker (Notice "Rings")
cap1-whisker1-rings01.jpg (445853 bytes)
Whisker "#1" - Close Up of "Rings"
cap1-whisker2-overall02.jpg (428171 bytes)
Whisker "#2" - Overall View
cap1-whisker2-base.jpg (460448 bytes)
Whisker "#2" - Base of Whisker
cap1-whisker2-mid-kink.jpg (375507 bytes)
Whisker "#2" - Kink Near Base (Notice "Rings")
cap1-whisker2-tip02.jpg (450600 bytes)
Whisker "#2" - Tip of Whisker (Notice "Rings")
cap1-whisker2-tip03.jpg (440509 bytes)
Whisker "#2" - Close Up of Tip (Notice "Rings")
cap1-whisker2-mid-rings01.jpg (370631 bytes)
Whisker "#2" - Middle of Whisker (Notice "Rings")

December 2001:  ~ 6 to 8 Months of Ambient Storage Have Elapsed Since the Thermal Cycle Exposure
The sample hybrid (with chip caps still mounted) has been stored at ambient condition ever since being received at GSFC.
SEM inspection in December 2001 shows that some of the tin whiskers on the chip caps have grown substantially since the previous inspection.  A few whiskers are now greater than 200 microns long (some approaching 250 microns). Previous examination in August 2001 showed maximum whisker length was approximately 100 microns.
Capacitors will be retained for additional ambient storage and inspection

cap1-04.jpg (35379 bytes) cap1-10.jpg (30552 bytes) cap1-13.jpg (29191 bytes)
cap1-14.jpg (35157 bytes) cap1-15.jpg (34996 bytes) cap2-03.jpg (52190 bytes)
cap3-03.jpg (32216 bytes) cap5-03.jpg (47063 bytes) cap5-04.jpg (43686 bytes)

August 2001 - NASA Goddard Inspection to Confirm Observations Reported by Hybrid Manufacturer

Q1.  What is the max length of whisker and typical length of whisker?  Quantify typical length of whiskers.
A1.  The max length of whisker was estimated to be 100 um and the typical length is estimated to be around 25um to 50 um.  This cap is populated with mainly two types of whiskers: toothpaste like (thick curly type) and needle like.  The density of the whiskers (including both types) is estimated to be 800 whiskers/mm2 (conservative estimate) which can be compared to GSFC experiment using pure tin plated brass substrates with ~50 whiskers/mm2 (max density sample).  Among entire whisker population, 5% is needle like whiskers that would grow potentially dangerously long.

Photos Courtesy of NASA Goddard Space Flight Center

Q2. EDAX on the cap termination to confirm if there's any lead (Pb) content?
A2. EDAX results showed Sn element exclusively.  No Pb was observed (Accuracy: Elapsed time was 500 sec.).

Q3.  Examine whisker growth at the conductive epoxy interface, and WITHIN the conductive epoxy?
A3.  The whiskers are present only on plated termination on both sides of the caps.  No whiskers were observable at the conductive epoxy interface nor within the conductive epoxy.

Q4.  Analyze the grain structure of the plating (quantify grain size) if possible.
A4.  This could not be done with SEM. Future Work will attempt to analyze the grain structure using Oriented Imaging Microscope (OIM) for grain structure analysis

February 2001-Photos Below Were Taken Shortly After 200+ Temp Cycles from -40C to +90C
Courtesy of I. Hernefjord


Previously Published Research on MLCCs and Tin Whiskers:

Prior to 2002 published research has suggested that multilayer ceramic capacitors (MLCCs) are "immune" to the risks of whisker formation.  Research by Murata** (M. Endo 1997) showed 18 years of whisker-free performance of pure tin plated MLCCs when stored continuously at 50C. In summary, the rationale given for this protection included:
Use of Nickel barrier metallization under the tin electroplate
Use of "Matte" tin plating chemistries and processes
Relatively "large" (>5 um), well-polygonized tin grains
Post electroplating annealing processes

** M. Endo, S. Higuchi, Y. Tokuda, and Y. Sakabe, "Elimination of Whisker Growth on Tin Plated Electrodes", Proceedings of the 23rd International Symposium for Testing and Failure Analysis, pp. 305 - 311, October 27-31, 1997

Responsible NASA Officials:

   Michael Sampson/NASA GSFC Code 306
   Dr. Henning Leidecker/NASA GSFC Code 562
Additional Researchers: 

   Jong Kadesch/Orbital Sciences Corp.
   Jay Brusse/QSS Group, Inc.

Last Updated:

April 20, 2005

This website provides information about tin whiskers and related research. The independent research performed during the past 50+ years is so vast that it is impractical to cover all aspects of tin whiskers in this one resource. Therefore, the absence of information in this website about a particular aspect of tin whiskers should NOT be construed as evidence of absence.

This material is presented to ensure timely dissemination of scholarly and technical work. For technical references posted herein, copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.